Real time automated three-dimensional recognition of micro/nano biological organisms

We present an overview of optical imaging system for 3D visualization and recognition of micro/nano biological organisms. For 3D sensing of a biological specimen, the diffraction pattern of the specimen is recorded on charge coupled device (CCD) image sensor. The recorded hologram is then transferred to the computer where 3D images of the specimen at different depths along the longitudinal direction are numerically reconstructed by using inverse Fresnel transformation. For 3D recognition and identification of micro/nano biological organisms, image segmentation algorithms are performed to identify regions of interest for further processing and then statistical classifiers using maximum likelihood estimator are used to classify the detected specimen into one of previously trained classes. It is shown in the experiments that the proposed system is useful for 3D sensing, recognition and classification of different biological specimen.